Mian M. Awais

Predicting weather events using fuzzy rule based system

Discovering and understanding the dynamic phenomena of weather to accurately predict different weather events has been an integral component of scientific investigations worldwide. The weather data, being inherently fuzzy in nature, requires highly complex processing based on human observations, satellite photography, or radar followed by computer simulations. This is further combined with an understanding of the principles of global and local weather dynamics. This paper attempts to solve weather event prediction for Lahore by implementing a fuzzy rule based system. The difficult problem of weather event prediction has been dealt in this paper through two separate experimental settings. In the first experimental setting a smaller dataset consisting of 365 instances with 4 inputs and 8 weather events has been used to develop a fuzzy inference system. In the second experimental setting the developed fuzzy system has been enhanced for a larger dataset consisting of over 2500 data points, having 17 inputs, and 10 weather events. For the later experiments the results of the fuzzy system have been compared with two other models i.e., decision tree (DT) based model and partial least square based regression (PLSR) model. It has been observed in the present study that the performance of the fuzzy system is sensitive to bootstrapping sampling technique that has been used for generating training and test samples for developing the fuzzy, DT and PLSR models. Further the models under consideration have been less sensitive to principal component analysis based dimensionality reduction method.

Enabling Self-Configuration in Autonomic Systems Using Case-Based Reasoning with Improved Efficiency

Autonomic computing is an emerging philosophy which promises to enable self-management capabilities in software systems. These self-management properties include self-configuration, self-healing, self-protection, self-optimization, self-awareness and self-governance. Enabling any of these properties in software systems is an open challenge. Exhibiting such self-management behavior is a continuous process in the software life cycle. Case-based reasoning is a problem solving methodology which exploits past experience. Past experience is maintained in the form of problem-solution pairs, also called cases. On the arrival of new problem, solution of past similar problems is used after appropriate adaptation. This problem solving technique can be used to achieve some of the properties of autonomic systems based on experience. To find this solution, entire experience space is searched which reduces efficiency. To overcome this efficiency problem, we restrict the fast growth of case repository, so that every time we have to search a very limited number of cases. We applied the proposed approach on a simulation of Autonomic Forest Fire application for self-configuration capability. Our results show that the proposed approach is quite promising in terms of accuracy as well as efficiency.

Application of internal model control methods to industrial combustion

Most practical systems are inherently non-linear to some extent in their behaviour and for their cost effective, smooth and safe operation, optimised control systems based on the non-linear models are required. To this end many useful techniques such as the stochastic modelling, sliding mode control and adaptive identification and control have been proposed in the literature. However, the high cost of implementation, the inability to capture imprecision with the required level of tolerance, and the in-flexibility against distortions in the operating variables, make them less attractive. To this end new artificial intelligence based techniques such as fuzzy logic, neural networks and probabilistic reasoning, are becoming more and more popular. Among these techniques neural networks have an edge over the others, mainly because of their ability to process large amount of available data, subsequent to the development of some interpretable models for solving engineering problems. Moreover, the ability to capture the non-linearities of a real system accurately and the versatility in being able to accommodate with ease, the various conventional and advanced strategies within their structures, make them much more attractive. The problem becomes more computationally worse and uncontrollable when inverse of the system does not exist. This problem is resolved when neural network based techniques such as internal model control (IMC) are applied to the real systems. This paper outlines the application of neural networks based IMC methods for estimation/control of important input and output variables of a 0.5MW laboratory scale industrial furnace. The application involves inputs such as the airflow rate, swirl number and momentum ratio. The outputs include emission levels of oxides of nitrogen especially nitric oxide. The response to step and staircase inputs has been analysed. The results have been compared with standard linear quadratic controller. The control output of the IMC methods has resulted in almost similar steady state error performance to the linear quadratic regulator. Although the development process of the IMC method might take longer time because of the training and data arrangement but has the capability of readjustment after being developed.

Achieving Self-configuration Capability in Autonomic Systems Using Case-Based Reasoning with a New Similarity Measure

A lot of activities inside human body are carried out intelligently without the explicit intervention of human itself, e.g. various actions of nervous systems, blood circulation system etc. Inspired from these natural systems, autonomic computing is an emerging concept which promises to enable such kind of self-management capabilities inside software systems. Case-based reasoning (CBR) is a methodology to solve current problems using the solutions of past problems of the similar nature. In this paper, we propose to use CBR to achieve self-configuration in autonomic systems. We introduce a new similarity measure to find nearest neighbors. We have also suggested the case preparation, case retrieval and case reuse and refinement methods to enable self-configuration in autonomic systems. To support our proposed methodology, we illustrate a case-study of Autonomic Forest Fire Application.

AdaBoost-based artificial neural network learning

A boosting-based method of learning a feed-forward artificial neural network (ANN) with a single layer of hidden neurons and a single output neuron is presented. Initially, an algorithm called Boostron is described that learns a single-layer perceptron using AdaBoost and decision stumps. It is then extended to learn weights of a neural network with a single hidden layer of linear neurons. Finally, a novel method is introduced to incorporate non-linear activation functions in artificial neural network learning. The proposed method uses series representation to approximate non-linearity of activation functions, learns the coefficients of nonlinear terms by AdaBoost. It adapts the network parameters by a layer-wise iterative traversal of neurons and an appropriate reduction of the problem. A detailed performances comparison of various neural network models learned the proposed methods and those learned using the least mean squared learning (LMS) and the resilient back-propagation (RPROP) is provided in this paper. Several favorable results are reported for 17 synthetic and real-world datasets with different degrees of difficulties for both binary and multi-class problems.

Finding focused itemsets from software defect data

Software product measures have been widely used to predict software defects. Though these measures help develop good classification models, studies propose that relationship between software measures and defects still needs to be investigated. This paper investigates the relationship between software measures and the defect prone modules by studying associations between the two. The paper identifies the critical ranges of the software measures that are strongly associated with defects across five datasets of PROMISE repository. The paper also identifies the ranges of the measures that do not necessarily contribute towards defects. These results are supported by information gain based ranking of software measures.

Autonomic Success in Database Management Systems

One of the primary uses of computer is to reduce cost and manage complexity with increase in efficiency and performance. Now system complexity is reaching a level that is beyond human ability. With the development of technology, people want to manage complex systems in an efficient and reliable manner. Development of raw computing power and proliferation of computer devices and usage of internet has grown up to exponential rates. This growth and unprecedented levels of complexity is leading towards new direction - Autonomic Computing. Autonomic features in system increase speed, efficiency, reliability and accuracy with less or no human interaction, ultimately providing error free environment. These autonomic capabilities are important in Database Management Systems (DBMSs). The DBMSs which have the capability to manage and maintain themselves are called Autonomic Database Management Systems (ADBMS). The ADBMSs are evolving from last many years. At present most of the activities in DBMS are performed autonomically and have achieved certain level of autonomicity. The paper identified some autonomic shortcomings in commercial DBMSs up to 2002. We made a survey on achievements of autonomic computing against these shortcomings in current DBMSs. For this purpose, we have studied and analyzed IBM DB2, Oracle and Microsoft SQL Server.

An Effort-Based Approach to Measure Completeness of an Entity-Relationship Model

For developing good quality information systems, the need of developing good quality conceptual models cannot be over emphasized. Completeness is an important characteristic that determines the quality of a conceptual model. To measure completeness, metrics called Completeness Index (CI) and Fuzzy Completeness Index (FCI) have been proposed in research. These metrics correspond to the extent functional dependencies are represented in a conceptual model, hi this paper, we propose an approach that takes into account the effort required to make a conceptual model complete. The higher the effort, the less is the degree of completeness. A metric Effort-based Completeness Index ECI is proposed accordingly. Although metrics previously proposed have an implicit relationship with the required effort, ECI provides the basis to explicitly relate completeness with the effort. Also, the method of computing ECI is simpler. In the proposed approach, first the effort based upon number of changes required to introduce a functional dependency in a given model is computed. Then the total effort for all functional dependencies in the problem domain determines ECI. hi this paper, ECI is calculated for different conceptual models and compared with corresponding CI and FCI values. It is illustrated that the proposed approach can easily be applied to practical problems in which alternative designs of the same problem are to be compared objectively in order to identify the one having better quality.

Dimensionally reduced Krylov subspace model reduction for large scale systems

Abstract This paper introduces a new mathematical approach that combines the concepts of dimensional reduction and Krylov subspace techniques for use in the model reduction problem for large-scale systems. Krylov subspace methods for model reduction uses the Arnoldi algorithm in order to construct the bases for controllability, observability, and oblique subspaces of state space realization. The newly developed algorithm uses principal component analysis along with Krylov oblique projection model reduction technique to provide computationally efficient and inexpensive model reduction method. To demonstrate the effectiveness of the proposed hybrid scheme the residual error, forward error and stability response analyses have been performed for various randomly generated large-scale systems.

Incomplete preference relations: An upper bound condition

In decision making, consistency in fuzzy preference relations is associated with the study of transitivity property. While using additive consistency property to complete incomplete preference relations, the preference values found may lie outside the interval [0, 1] or the resultant relation may itself be inconsistent. This paper proposes a method that avoids inconsistency and completes an incomplete preference relation using an upper bound condition. Additionally, the paper extends the upper bound condition for multiplicative reciprocal preference relations. The proposed methods ensure that if n-1 preference values are provided by an expert, such that they satisfy the upper bound condition, then the preference relation is completed such that the estimated values lie inside the unit interval [0, 1] in the case of preference relations and [1/9, 9] in the case of multiplicative preference relation. Moreover, the resultant preference relation obtained using the proposed method is transitive.